1 Application Performance Testing for Data Centre Relocation
2 Latency and Data Centre Relocation Background Companies across the world are moving their applications and data servers from centralised servers to remote data centres, consolidating those services and utilising cloud providers to outsource large sections of their IT infrastructure. These strategies offer significant benefits. For example, IT costs are significantly reduced, business continuity and information security are improved, service management is optimised, and government and industry regulations complied with. Network-related delays can change dramatically during data centre relocation, since geographic distance contributes significantly to the network latency between client and server. Take for example the simple doubling of latency (as the server is now physically further away from the user). At face value, this may appear that operations will take twice as long, but increased latency also reduces the throughput of applications, further impacting end users. The relationship between network delay and application response time is far from one-to-one. The following chart provides a relatively simple example of the relationship between network latency and application latency. According to analysts, 88% of enterprises are currently in the early phases of either consolidating servers or executing major data centre relocation. The Challenge With so many benefits come so many challenges. According to Gartner, one of the primary factors inhibiting these advantageous projects is the concern over how business critical applications and services will be adversely affected by such moves. Such concerns are legitimate. Consolidations and relocations have a huge impact on application performance and service levels in many ways. Users that were previously local to the servers supporting their business applications, for example, may become remote users. Few applications are future-proofed against these types of network changes. Varying bandwidth, jittery links, increased latencies and fluxing service quality must always be considered prior to a data centre migration project. For a local user with less than a millisecond of network delay between the client and the server, the transaction completed in three seconds. When just 50 milliseconds of network delay was introduced (representing a typical cross-country WAN connection), the performance of this transaction did not slow down by only 50 milliseconds. Instead, this same login transaction took a full 30 seconds to complete. This example dramatically illustrates how small changes in network latency result in major problems with application performance. In a complex project such as a data centre move, the challenge of accessing the network impact on application performance becomes one of large scale as hundreds of servers hosting thousands of applications need to be tested for the impact of the data centre move.
3 Testing and Delivering the Data Centre Relocation There are many issues that can arise during a data centre move. These issues can threaten business operations and undermine the credibility of IT. Many of these issues cannot be predicted by IT organisations that operate in traditional silos, because they re caused by subtle, complex interactions between networks, servers and applications. In fact, after these issues manifest themselves in the port-move production environment, many IT organisations are left struggling to understand and resolve them. There are a number of options for an organisation involved in data centre relocation choosing the right option is crucial to the business. Option one: Performing the Move Immediately This method involves organisations moving the live systems and their data to the new data centre over a down-time period (such as a weekend). This approach delivers the relocation without any testing and is primarily used when the cost of detailed analysis is prohibitive. Advantages: Delivers the relocation immediately Miniscule costs for relocation or testing Disadvantages: Exceptionally high risk Application performance will not be verified until an operation is performed (i.e. weekly or monthly tasks) Having to manage both local and relocated server(s) in case a fall back is required Duplication of end-users work if fall back is required Extensive hidden costs post deployment from fall backs or application failure Poor end-user experience and tarnishing of professional reputations Failure of SLA s
4 Option Two: Lab Testing and Research In the past, the most popular option has been to perform laboratory testing and research with slow migration of end-users to any newly relocated systems. This approach allows for extensive testing and an acceptable understanding of the expected end-user experience, but is an extremely costly operation. This form of testing requires servers and end-users to be duplicated in a controlled and separate environment from the live systems. Due to the extensive size of most systems only a small number of end-user systems are normally duplicated. Advantages Allows for controlled experiementling with no impact on live systems Small scale evaluation and analysis The report can be used to gauge applications and operations that have greater stress on servers Disadvantages Expense of servers, clients machines and hiring test equipment Significant cost of professionals to understand and details the characrteristics of applications Time required to generate detailed information on the application performance Limited scope of testing may result in incomplete risk analysis
5 Option Three: Live Performance Testing Live performance testing for a data centre relocation is used by implementing a WAN Emulator between the end-users and the live servers. In its default state, the WAN Emulator acts as a high performance router / switch resulting in zero down-time. The WAN emulator provides an organisation with the ability to slowly increase the latency and reduce bandwidth over a period of time. Users can actively report their experiences between each change. Realistic simulation of the data centre relocation Ability to simulate many difference remote sites (with different types of network connections) Significantly reduced costs with no external consultants or specialists Live statistics on network utilisation Built-in reporting that provides information when problems arise, allowing rapid and cost-effective resolution As this is a live solution you can immediately fall-back to the orginal unimpaired network with a single click No duplication of end-user during fall-back or post relocation Fortunately, a data centre move does not have to be a gamble. By applying proven best practices and appropriate simulation, testing and engineering techniques early in the process, IT organisations can bring predictability to their data centre moves. They can effectively simulate each stage of the move to ensure that the business is properly supported throughout the life of the project. Just as important, this disciplined approach of data centre and server relocation helps reduce costs, eliminate unexpected delays and optimize user acceptance.
6 Linkedin.com/company/jartechnologies How JAR can help JAR can assist Data Centre projects in a number of important ways: Pre-migration: Our emulators are used in the proof of concept and planning stages by providing what if scenario profiling of application performance from the proposed data centre. Our self-service WAN emulators can introduce latency, jitter, loss and many other network impairments found following a server or data centre relocation, so application performance issues can be identified and mitigated early, and SLA s met once the data centre is operational. On-going Data Centre Maintenance: As companies develop and/or deploy new and updated business critical applications, it is vital for DC stakeholders to understand exactly how these new software modules will react across the end-to-end network (from data centre to end user and back again). WAN acceleration technology will not solve the issue of poor application performance if the issue is not simply related to low available bandwidth, as fundamental issues with application code will not be resolved by sending the data faster. Trailing new or updated application traffic on production servers can quickly lead to interference with other live applications, and the network conditions experienced at any one time may be significantly different later the same day, week or month. Our WAN emulation solutions allow customers to test under the network conditions their applications will be expected to perform on and determine if, how and why applications will or will not perform when hosted from the data centre. JAR:Emulate series of network emulators enables you to: Reduce costs: Identifying and eliminating application/system performance issues early in the development lifecycle results in significant cost savings. These are primarily achieved through a dramatic reduction in the level of remedial work required later. Maintain reputations and retain customers: Companies who are unaware of how their applications/systems will perform when released on the WAN are prone to a measurably higher level of product recall and re-work in the field. Reduce time to market: Identifying and fixing application performance issues early in the development lifecycle will prevent release delays to users and customers. Ensure Service Level Targets are met: Know with certainty that your application or system will perform as required when released onto a WAN, and service level targets will be met. JAR Technologies Ltd Howard House, 1 Brunswick Street, Belfast, BT2 7GE, United Kingdom JAR Technologies Ltd 2-4 Wellington Street, Belfast, BT1 6HT, United Kingdom T: +44 (0) E: W:
Best Practices for VoIP Network Assurance and Performance Management 1-800-508-5233 PathView_Sales@AppNeta.com AppNeta.com White Paper Best Practices for VoIP Network Assurance & Performance Management
Taking the Risk out of VoIP Deployment and Ownership: Seven Best Practices for Convergence Success A Shunra Software White Paper Introduction Voice-over-IP (VoIP) technology offers a wide range of benefits
3 NREN and its Users The NREN s core activities are in providing network and associated services to its user community that usually comprises: Higher education institutions and possibly other levels of
MOVING THE DATA CENTER FROM CHAOS TO CONTROL BEST PRACTICES IN DATA CENTER INFRASTRUCTURE MANAGEMENT WITH INTEGRATED PROCESSES AND TECHNOLOGIES WHITE PAPER 101 TABLE OF CONTENTS 01 EXECUTIVE SUMMARY 03
HP Performance Engineering Best Practices Series for Performance Engineers and Managers Performance Monitoring Best Practices Document Release Date: May 2009 Software Release Date: May 2009 Legal Notices
HP Performance Engineering Best Practices Series for Performance Engineers and Managers Performance Monitoring Best Practices Document Release Date: 201 Software Release Date: 2014 Legal Notices Warranty
Does My Infrastructure Look Big In This? A review of Fujitsu s Infrastructure Services Author: Robin Bloor Published: April 2002 Bloor Research Table of Contents Contents Management Summary................................
Migration Planning Kit Microsoft Windows Server 2003 This educational kit is intended for IT administrators, architects, and IT managers. The kit covers the reasons and process you should consider when
VoIP Lifecycle Management Best Practices and Business Value Viola Networks May 2006 VoIP Lifecycle Management VoIP Lifecycle Management Best Practices and Business Value When managing an enterprise VoIP
Technical Considerations in a Windows Server Environment INTRODUCTION Cloud computing has changed the economics of disaster recovery and business continuity options. Accordingly, it is time many organizations
Outsourcing Network Support: The Surprising Strategy That Helps You Spend Less for Higher Uptime How small and medium-sized businesses (SMBs) are outsourcing network support to reduce spending, improve
Exploiting the Experience of Transformation IT Outsourcing 2006 IT World Limited on behalf of the BuyIT Best Practice Network Page 1 P12 IT Outsourcing May 2006 Forewords One of the prime objectives of
WHY OPERATIONAL MANAGEMENT IN VIRTUALIZED DATA CENTERS IS CRITICAL TO DELIVERING SERVICE ASSURANCE Most large enterprises' business processes require IT to function smoothly and efficiently, making a robust
Cloud Computing Payback An explanation of where the ROI comes from November, 2009 Richard Mayo Senior Market Manager Cloud Computing email@example.com Charles Perng IBM T.J. Watson Research Center firstname.lastname@example.org
Firewall Strategies June 2003 (Updated May 2009) 1 Table of Content Executive Summary...4 Brief survey of firewall concepts...4 What is the problem?...4 What is a firewall?...4 What skills are necessary
Avoid Three Common Pitfalls With VoIP Readiness Assessments xo.com Table of Contents Abstract..................................................... 1 Overview.....................................................
White Paper An Enterprise Security Program and Architecture to Support Business Drivers seccuris.com (866) 644-8442 Contents Introduction... 3 Information Assurance... 4 Sherwood Applied Business Security
FRAUNHOFER RESEARCH INSTITUTION AISEC CLOUD COMPUTING SECURITY PROTECTION GOALS.TAXONOMY.MARKET REVIEW. DR. WERNER STREITBERGER, ANGELIKA RUPPEL 02/2010 Parkring 4 D-85748 Garching b. München Tel.: +49
Cisco SAFE: A Security Reference Architecture The Changing Network and Security Landscape The past several years have seen tremendous changes in the network, both in the kinds of devices being deployed
WHITE PAPER Introduction... 2 Reduce Tool and Process Sprawl... 2 Control Virtual Server Sprawl... 3 Effectively Manage Network Stress... 4 Reliably Deliver Application Services... 5 Comprehensively Manage
Cost-Effective Alternatives to Software Asset Management kpmg.com Contents Executive Summary 1 Introduction 2 Key SAM issues 4 A cost-effective approach to SAM 6 Benefits of SAM 8 Conclusion 9 Cost-Effective
Assess, Adjust, Improve An LXI Publication Page 1 of 11 Your company's ability to recover is a high priority. In a survey by Contingency Planning & Management Magazine of 1437 contingency planners, 76%
fs viewpoint www.pwc.com/fsi July 2013 02 14 17 20 27 31 Point of view Competitive intelligence A deeper dive into the benefits of cloud A framework for response How PwC can help Appendix Clouds in the
Brand icons C O M M U N I C AT I O N S, S O RT E D. Cloud Connectivity Inbound Mobile VoIP Connectivity Inbound Mobile VoIP Inbound Mobile VoIP Brand icons - reversed Cloud Brand icons - green background
Issue 4 Handling Inactive Data Efficiently 1 Editor s Note 3 Does this mean long term backup? NOTE FROM THE EDITOR S DESK: 4 Key benefits of archiving the data? 5 Does archiving file servers help? 6 Managing
: Ensuring End-to-End Service Quality and Performance in a Multi-Vendor Environment A Executive Summary Creating Business-Class VoIP Traditional voice services have always been reliable, available, and